Fallacies that Misinform Our Thinking About Michigan’s Population … · 2016. 2. 26. · 1 Kenneth Darga, State Demographer Michigan Revenue Estimating Conference January 11, 2008

1

Kenneth Darga, State Demographer

Michigan Revenue Estimating Conference

January 11, 2008

Fallacies that Misinform Our Thinking About Michigan’s Population and Economy

The map on the title slide was prepared in 1831 by David H. Burr of New York. It has been made available by the Michigan State University Library

______________________________________________________________

Thank you for the opportunity to address this year’s Revenue Estimating Conference.As I was looking over my notes last night, I realized that the title of my presentation is ambiguous. The title describes the subject matter of my remarks--not their content. Just so there is no confusion, I want to make it clear that the purpose of this presentation is to discuss and correct fallacies, not to propagate new ones.There are several common beliefs about Michigan’s population and economy that are not quite true. All five of the fallacies that I will talk about this morning have elements of truth—there is valid evidence that seems to support them—but they can misinform our thinking about Michigan’s economy in some important ways.Each of these fallacies also has an important bearing on where Michigan has been and where it is going. Our state is certainly not where it wants to be economically, but I hope you will excuse me if a few of the things that I say about Michigan this morning are not as negative as you might expect..

2

Fallacy #1:

Michigan has a chronic problem of net out-migration.

Ever since I started working as a demographer for the State of Michigan in 1991, there have been newspaper articles and other reports about massive migration out of Michigan.

3

Total Net Migration: Michigan, 1960-2005

-200,000

-150,000

-100,000

-50,000

0

50,000

100,000

60-6162-6364-6566-6768-6970-7172-7374-7576-7778-7980-8182-8384-8586-8788-8990-9192-9394-9596-9798-9900-0102-0304-05

Interval

Num

ber

of P

erso

ns

Data Source: Pop. estimates from U.S. Census BureauVital statistics from Mich. Dept.of Community Health

Based on population estimates from the Census Bureau, I started telling audiences a few years ago that in-migration and out-migration from Michigan had been roughly in balance since about 1985.

In this slide, the data points above the black line represent years with more in-migration than out-migration, and the data points below the line represent years with more out-migration. Over a 20 year period, there were 10 years with a little bit of net in-migration and 10 years with a little bit of net out-migration.

4


-200,000

-150,000

-100,000

-50,000

0

50,000

100,000

60-61

63-64

66-67

69-70

72-73

75-76

78-79

81-82

84-85

87-88

90-91

93-94

96-97

99-00

02-03

05-06

Interval

Num

ber

of P

erso

ns


Last year, I had to modify that message.

I still said that in-migration and out-migration had been roughly in balance since about 1985, but I also pointed out that survey data for 2005 and population estimates for 2005 and 2006 were less favorable than the figures for previous years.

5


-200,000

-150,000

-100,000

-50,000

0

50,000

100,000

60-6162-6364-6566-6768-6970-7172-7374-7576-7778-7980-8182-8384-8586-8788-8990-9192-9394-9596-9798-9900-0102-0304-0506-07

Interval

Num

ber

of P

erso

ns


Based on downward revisions of the old data, as well as new data that has been released in the past few months, I am revising my message again for 2008: Michigan experienced significant net out-migration from 2005 through 2007, but it is not a chronic pattern.

Michigan formed an image of itself in the 1980’s as a state that loses a lot of population through migration. Even through twenty years without significant net out-migration, some analysts and commentators never managed to shake that self-image. Despite the trend over the past three years, Michigan does not have chronic out-migration.

6

Fallacy #2:

Michigan has a chronic pattern of high out-migration by young people.

But what about young people?Fallacy #2 is that Michigan has chronic pattern of high out-migration by young people.

7

Migration to Other States by Age: 2000-06 for the United States

0%

1%

2%

3%

4%

5%

6%

0 10 20 30 40 50 60 70 80

Age

Mig

ratio

n R

ate

Data Source: U.S. Census BureauAmerican Community Survey (IPUMS.org)

Here is the sort of data that this fallacy is based on. This chart shows the percentage of people in each age group who move to another state. The previous chart showed NET migration, but this chart shows out-migration by itself. I will talk about in-migration in a few minutes.

This data is from the Census Bureau’s American Community Survey, a very large sample survey that is taking the place of the Census Long Form.

You can see that out-migration rates are dramatically higher for young adults than for any other age group.

The important thing to know about this chart is that it is not based on Michigan data. This is the national pattern of inter-state migration by age. Every state has high out-migration by young adults, because young adults move around a lot.

8

Migration to Other States by Age for Michigan and the U.S.: 2000-2006

0%

1%

2%

3%

4%

5%

6%

0 10 20 30 40 50 60 70 80

Age

Mig

ratio

n R

ate

United States

Michigan (Out)


This red line represents Michigan’s pattern of out-migration by age. Michigan has actually had a lower rate of out-migration to other states than the rest of the country for every age group below age 65.

This chart covers the entire period from 2000 through 2006. Michigan did quite a bit better in the first 5 years of that time period than in the most recent 2 years, so I will show data for those two time periods separately.

9

Domestic Migration by Age for Michigan and the U.S.: 2000-2004

0%

1%

2%

3%

4%

5%

6%

0 10 20 30 40 50 60 70 80

Age

Mig

ratio

n R

ate

United StatesMichigan (Out)Michigan (In)


The red line in this chart shows out-migration for just the period from 2000 through 2004, and the blue line represents in-migration to Michigan by age.

This is what I think of as Michigan’s normal pattern of migration. This is almost exactly what Michigan’s migration patterns looked like from 1995 to 2000, according to the 2000 Census.

Our in-migration rate is highest for young adults, just like our out-migration rate is. And migration into Michigan from other states was just about exactly equal to migration out of Michigan for all the age groups below age 45. We even seemed to have small net gains fromother states for people between the ages of 26 and 34.

10


0%

1%

2%

3%

4%

5%

6%

0 10 20 30 40 50 60 70 80

Age

Mig

ratio

n R

ate



Michigan does tend to lose population to other states after age 45, and especially after age 65. But that population loss is generally offset by immigration of working-age people from other countries.

The previous set of charts showed that Michigan’s overall in-migration and out-migration were essentially in balance during this time period. That is because our loss of people in the older age groups was just about exactly offset by foreign immigration. Foreign immigration is not reflected in this chart. This chart shows just domestic migration.

11


0%

1%

2%

3%

4%

5%

6%

0 10 20 30 40 50 60 70 80

Age

Mig

ratio

n R

ate



Out-migration from Michigan went up in 2005 and 2006 for some age groups. The biggest increase was for people age 22-26. Yet, even in these years, our out-migration was not above the national rate.

What Michigan really stands out for in 2005 and 2006 is an extremely low rate of in-migration from other states, especially for people age 22-26.

Our migration pattern has definitely changed for the worse in the past few years, but it is our in-migration that is out of line with national rates. Our out-migration is not.

12

Fallacy #3:

Michigan has a chronic brain drain.

Thus, until the past few years, Michigan had a long period without significant net out-migration by people under the age of 45.But what about people with college degrees?

Fallacy #3 is that Michigan has a chronic brain drain.

13

Out-Migration by Educational Attainment for Persons Age 22-34: Michigan, 2000-2006

3.4%2.3% 2.1%

6.3%7.1%

0%

1%

2%

3%

4%

5%

6%

7%

8%

Total Associate's Degree Graduate/ProfessionalDegree

Ave

rage

Ann

ual R

ate

of O

ut-M

igra

tion

Data Source:

U.S. Census BureauAmerican Community Survey (IPUMS.org)

Total No Degree Associate's Bachelor's Grad./Prof.

Here is the kind of data that leads people to believe that their state has a brain drain.From 2000 through 2006, an average of 3.4 percent of Michigan’s residents between the ages of 22 and 34 moved from Michigan to other states.For people without degrees, the rate was only 2.3 percent.For people with Associate’s degrees: 2.1 percent.For people with Bachelor’s degrees: 6.3 percent.For people with Graduate and professional degrees: 7.1 percent.

It looks like Michigan must have a terrible brain drain.What many people overlook is that there is also a large flow of educated people INTO Michigan from other states. People with college degrees tend to move around more than people without degrees. They are more likely to do multi-state job searches.

14

Michigan, 2000-2006Migration by Educational Attainment for Persons Age 22-34

0

10,000

20,000

30,000

40,000

50,000

NoDegree Assoc Bach Grad/Prof All Degrees

Ave

rage

Ann

ual N

umbe

r of

Mig

rant

s

OutIn

No Degree Associate's Bachelor's Graduate/Prof. All Degree Degree Degree Degrees

52 : 48

49: 51

52 : 48

47 : 53

50 : 50


Since the 2000 Census, Michigan has actually gained about as many brains from other states as it has lost.

Once again, I will show data for the entire 7 year period first, and then for the periods “2000 through 2004” and “2005 through 2006.”Over the entire 7 year period, Michigan has been close to balance for people in this age group without degrees. 52 left Michigan for every 48 who came to Michigan from other states.

We gained slightly more young people with associate’s degrees than we lost. We lost slightly more people with bachelor’s degrees than we gained. Some went to graduate schools or professional schools in other states. And we gained more people with graduate and professional degrees than we lost. Some were Michigan natives coming back home.

For all degrees put together, our ratio is 50:50. We gained about as many people with degrees as we lost. Michigan also receives educated immigrants from other countries, who are not reflected in this chart.

15


0

10,000

20,000

30,000

40,000

50,000

1 2 3 4 5

Ave

rage

Ann

ual N

umbe

r of

Mig

rant

s

OutIn


50 : 50

38 : 62

49 : 51

46 : 54

40 : 60

Once again, Michigan did better from 2000-2004 than in the two more recent years.

We actually gained quite a few more educated residents than we lost during that period. You never would have suspected that from articles you may have read in the newspapers, but that is what actually happened, according to the Census Bureau’s American Community Survey. Part of our “brain gain” represented Michigan natives returning from colleges and universities in other states.

16


0

10,000

20,000

30,000

40,000

50,000

1 2 3 4 5

Ave

rage

Ann

ual N

umbe

r of

Mig

rant

s

OutIn


57 : 43

68 : 32

59 : 41

63 : 37

61 : 39

However, Michigan did have a substantial brain drain in 2005 and 2006. We currently have a net brain drain, but it is a new phenomenon rather than a chronic pattern.

17

Fallacy #4:

Michigan has a chronic unemployment problem, and it currently has one of the highest unemployment rates in the nation’s recent history.

In summary, data from the Census Bureau shows that Michigan doesnot have a chronic migration problem, and that even our current pattern of migration is not as bad as it is often portrayed to be. Everybody isn’t really leaving Michigan.

But why not?

Fallacy #4: Michigan has a chronic unemployment problem, and itcurrently has one of the highest rates of unemployment in the nation’s recent history.

Michigan does have a serious unemployment problem, but there are some ways of overstating the problem that are not valid--and that statement is one of them.

18

Unemployment Rates: 1976-2007

0%

4%

8%

12%

16%

20%

19761977197819791980198119821983198419851986198719881989199019911992199319941995199619971998199920002001200220032004200520062007

Year

Une

mpl

oym

ent R

ate

Seasonally Adjusted Monthly Data for January 1976 through Novermber 2007Excludes Washington D.C. and Puerto Rico

Data Source: U.S. Bureau of Labor Statistics

This chart shows the nation’s unemployment rate since 1976. In November 2007—the latest figure currently available--the U.S. had one of its lowest rates of unemployment in the past 31 years, surpassed only by a 3-1/2 year period leading up to the 2001 recession, and a 17 monthperiod in 2006 and 2007.

19


0%

4%

8%

12%

16%

20%

19761977197819791980198119821983198419851986198719881989199019911992199319941995199619971998199920002001200220032004200520062007

Year

Une

mpl

oym

ent R

ate

U.S.

Michigan



Michigan, on the other hand, has had a higher unemployment rate than the U.S. as a whole for the past several years. Our rate of unemployment has stayed about the same since the 2001 recession,while unemployment went down in the remainder of the United States.

But there are a few important things that I would like to point out about Michigan’s unemployment rates.

20


0%

4%

8%

12%

16%

20%

19761977197819791980198119821983198419851986198719881989199019911992199319941995199619971998199920002001200220032004200520062007

Year

Une

mpl

oym

ent R

ate

U.S.

Michigan



First of all, Michigan recently had a ten year period in which its unemployment rate was close to or below the national rate. Our current unemployment gap is a fairly new problem.

21


0%

4%

8%

12%

16%

20%

19761977197819791980198119821983198419851986198719881989199019911992199319941995199619971998199920002001200220032004200520062007

Year

Une

mpl

oym

ent R

ate



Second, as high as our unemployment rate is, it is not the highest in our recent history.

Since January of 1976, Michigan’s unemployment rate has been higher than its current level 47 percent of the time.Sometimes, it was much higher.

Thus, one reason why our migration rates are not as bad as in other periods of our history is that our unemployment rate is not as high as in other periods.

22


0%

4%

8%

12%

16%

20%

19761977197819791980198119821983198419851986198719881989199019911992199319941995199619971998199920002001200220032004200520062007

Year

Une

mpl

oym

ent R

ate

U.S.MichiganLowest StateHighest State



Third, there has been an important positive development in the nation’s pattern of unemployment over the past couple decades.

The gray line toward the bottom of this chart shows the lowest state unemployment rate in each time period, and the gray line toward the top of the chart shows the highest state unemployment rate. The lowest rate is usually in a rural state, such as Nebraska or the Dakotas. Idaho is the state currently in the top spot. The highest rate is most often in West Virginia or Alaska.

Louisiana had the highest rate after Hurricane Katrina, and Michigan and several other states have also had turns at the top.

23


0%

4%

8%

12%

16%

20%

19761977197819791980198119821983198419851986198719881989199019911992199319941995199619971998199920002001200220032004200520062007

Year

Une

mpl

oym

ent R

ate



Michigan has been at the top for 17 of the past 20 months, and it was also at the top in the early 1980’s.

24


0%

4%

8%

12%

16%

20%

19761977197819791980198119821983198419851986198719881989199019911992199319941995199619971998199920002001200220032004200520062007

Year

Une

mpl

oym

ent R

ate

U.S.MichiganLowest StateHighest State



The nation’s lowest unemployment rate has not changed a great deal over the past few decades.But the nation’s highest rate has gone down considerably since 1983.

Back in the 1970s and 1980s, some states would typically have unemployment rates above 10%—and sometimes as high as 18%—while other states had labor shortages. But in more recent years, the nation’s highest unemployment rates have been around 7 or 8 percent.

This convergence between the nation’s highest and lowest rates is an important development. It is causing unemployment to have somewhat less impact on migration rates than in previous years.

25


0%

4%

8%

12%

16%

20%

19761977197819791980198119821983198419851986198719881989199019911992199319941995199619971998199920002001200220032004200520062007

Year

Une

mpl

oym

ent R

ate



Michigan’s most recent unemployment estimate, which is for November 2007, is much higher than we want it to be, but it is not very high compared to the nation’s highest rates in prior years.

52 percent of the time, the gap between Michigan’s rate and the lowest rate in the country has been higher than it is now, including several recent months. The gap is currently much smaller than it was in the early 1980’s.

26


0%

4%

8%

12%

16%

20%

19761977197819791980198119821983198419851986198719881989199019911992199319941995199619971998199920002001200220032004200520062007

Year

Une

mpl

oym

ent R

ate



And 77 percent of the time since this data series began in 1976, the nation’s highest level of unemployment exceeded Michigan’s most recent rate.

Thus, one reason why Michigan has not had an extreme level of out-migration in recent years is that its unemployment rate has not been extreme by historic standards.

27

• Estimated unemployment rates for November, 2007:– 7.4% Michigan– 4.7% United States– 2.7% Lowest Rate (Idaho)

• Since the current data series began in January of 1976:– 47% of the time, Michigan’s unemployment rate has been higher

than its current level (highest: 16.9%, November 1982)– 15% of the time, Michigan has had the nation’s highest rate,

including 17 of the past 20 months and 30 months in the late 70’s and early 80’s.

– 52% of the time, the gap between Michigan’s rate and the lowest rate in the country has been higher than Michigan’s current gap.

– 77% of the time, the nation’s highest level of unemployment has been higher than Michigan’s most recent rate. (highest: 18.2% in West Virginia, March 1983)

28

Fallacy #5:

Not much hiring takes place during difficult economic times.

Fallacy #5: Not much hiring takes place during difficult economic times.

That is the impression that you might get sometimes from reading the newspaper or watching the news on TV. Anecdotal accounts in thenews media can play a very powerful role in building hope or despair about Michigan’s economy.

Some media outlets focus on Michigan success stories, and their reports can be very encouraging. And some media outlets focus primarily on stories about people leaving Michigan and people losing their jobs. Those reports can be very discouraging.

29

It is possible to find dozens of negative anecdotes—or even hundreds or thousands of negative anecdotes—regardless of whether a state is having a good year or a bad year.

(And the same can be said for positive anecdotes.)

Whether you are writing reports about Michigan’s economy or reading those reports, it is important to be aware of a statistical reality that is often overlooked: It is possible to find dozens of negative anecdotes—or even hundreds or thousands of negative anecdotes—regardless of whether a state is having a good year or a bad year. And, of course, the same thing can be said about positive anecdotes.

30

To provide a statistically balanced picture of Michigan’s job market in a difficult year like 2006, a reporter or analyst would present 9 examples of people finding jobs in new and expanding companies for every 10examples of people losing jobs in companies that close or cut back.

To provide a statistically balanced picture of Michigan’s job market in a difficult year like 2006, a reporter or analyst would present 9 examples of people finding jobs in new or expanding companies for every 10 examples of people losing jobs in companies that close or cut back.

That would reflect what the Bureau of Labor Statistics says actually happened in 2006.

31

Private-Sector Job Loss and Job Gain in Michigan: 2006

994 K 927 K

0 K

400 K

800 K

1200 KN

umbe

r of

Job

s

Job Loss in Firms that Close or Contract

Job Gain in FirmsEstablished or Expanded

Data Source: U.S. Bureau of Labor Statistics, Business Employment Dynamics

This chart is based on a new series of data that was released a few months ago by the U.S. Bureau of Labor Statistics. This data series is called “Business Employment Dynamics,” and it is based on the quarterly payroll reports that are filed by all employers covered by the Unemployment Insurance system. This dataset covers 98 percent of the non-farm civilian labor force.*

It shows that 927,000 jobs were gained in Michigan firms being established or expanded in 2006, while 994,000 jobs were lost in firms that contracted or closed.

That is actually not a very good ratio, but it is not nearly as bad as it is often portrayed to be. I have very seldom seen articles about Michigan’s job market that have 9 examples of people hired in expanding firms for every 10 examples of people losing jobs in firms that are closing or cutting back.

*Major exclusions: self-employed workers, religious organizations, most agricultural workers on small farms, members of Armed Forces, elected officials in most states, most employees of railroads, some domestic workers, most student workers at schools, and employees of certain nonprofit organizations.

32


0

10,000

20,000

30,000

40,000

50,000

1 2 3 4 5

Ave

rage

Ann

ual N

umbe

r of

Mig

rant

s

OutIn


57 : 43

68 : 32

59 : 41

63 : 37

61 : 39

The same statistical reality applies to other topics.

To provide a statistically balanced picture of Michigan’s brain drain in a difficult year like 2005 or 2006, a reporter or analyst would provide 4 examples of young college graduates coming to Michigan for every 6 accounts of young college graduates leaving Michigan.

In fact, that is the same ratio that would be used for young people without degrees.

Again, that is not a very favorable ratio, but it is not nearly as bad as it is often portrayed to be.

33

Analysts and commentators and reporters have to decide what story they will tell, and that is often done months or years before the relevant data becomes available.

If they decide that the story is that all the jobs or all the people are leaving Michigan—and if they choose anecdotal evidence accordingly—they can lead readers or listeners to believe damaging fallacies about Michigan’s population and economy.

Analysts and commentators and reporters have to decide what story they will tell. And that is often done months or years before the relevant data becomes available.

If they decide the story is that all the jobs or all the people are leaving Michigan—and if they choose anecdotal evidence accordingly—then readers or listeners can be led to believe damaging fallacies about Michigan’s population and its economy.

Documents

Fallacies that Misinform Our Thinking About Michigan’s Population … · 2016. 2. 26. · 1 Kenneth Darga, State Demographer Michigan Revenue Estimating Conference January 11, 2008